Multi-input fuzzy inference system based model to predict the cutting temperature when milling AISI 1060 steel

The increase in the cutting temperature during milling has harmful effects which negatively affect the technical and economic machining characteristics such as: residual stresses, dimensions of machined parts and tools life. The nature of milling operations and the tool geometry make it difficult to predict or measure the temperature during the machining process, which is why great attention has been paid to measurement and prediction methodologies of cutting temperature during milling. In this work, a new intelligent identification technique of the cutting temperature based on the fuzzy set theory has been proposed to replace the strategy based on the operator qualification. This technique uses a fuzzy multiple input inference system to determine the influence of the cutting parameters on the cutting temperature. The fuzzy modeling is based on an experimental database resulting from the non-contact measurement of cutting temperature using an infrared camera with an emissivity setting adapted to the material. The results of the fuzzy system show that the fuzzy model is able to specify results providing a very good correlation between the experimental data and those predicted. The average error of the model was approximately 2.242%. The parameters used for the validation of the model were different from the data used for the construction of the fuzzy rules. The results showed that the most important parameter on the cutting temperature is depth of cut. The results obtained in this paper show that the developed model can be applied to predict the cutting temperature with precision during the milling process.